model calculations multi-platform observations and Atmospheric Bromine Monoxide:

Texte intégral

(1)

UNIVERSITE LIBRE DE BRUXELLES

Faculté des Sciences appliquées

Année académique 2009 - 2010

Atmospheric Bromine Monoxide:

multi-platform observations and model calculations

Directeur de Thèse : Michel Herman Dissertation présentée par Nicolas THEYS en vue de l’obtention du grade de Co-promoteur : Michel Van Roozendael

Docteur en Sciences de l’Ingénieur

(2)

Cover :

Looking through the Earth's atmosphere from on board the International Space Station (ISS).

Credits: European Space Agency (ESA)

(3)

UNIVERSITE LIBRE DE BRUXELLES

Faculté des Sciences appliquées

Année académique 2009 - 2010

Atmospheric Bromine Monoxide:

multi-platform observations and model calculations

Directeur de Thèse : Michel Herman Dissertation présentée par Nicolas THEYS en vue de l’obtention du grade de Co-promoteur : Michel Van Roozendael

Docteur en Sciences de l’Ingénieur

(4)
(5)

Remerciements

Le travail présenté dans cette thèse a été réalisé à l’Institut d’Aéronomie Spatiale de Belgique (IASB) à Bruxelles.

Je tiens à remercier chaleureusement Michel Van Roozendael, mon promoteur de thèse à l’IASB, pour la confiance qu’il m’a accordée au cours de ces dernières années. La porte de son bureau a toujours été grande ouverte pour un conseil ou une discussion. Il n’a jamais compté les heures qu’il m’a consacrées, malgré la pile de tâches urgentes qui l’attendaient parfois. Son œil critique et son enthousiasme ont toujours été un moteur important pour mon travail et pour la rédaction de cette thèse. Mes remerciements vont aussi à Michel Herman, mon directeur de thèse à l’Université Libre de Bruxelles, pour sa disponibilité, ainsi qu’aux membres de mon comité d’accompagnement, Michel Carleer et Michel Godefroid.

Je voudrais aussi particulièrement remercier François Hendrick, pour sa patience et sa gentillesse. Son soutien sans faille et nos discussions ont toujours été constructives pour mon travail de recherche. Je le remercie également pour la relecture minutieuse de ce manuscrit et de mes publications.

Un grand merci à mes chers collègues de bureau Isabelle De Smedt, Gaïa Pinardi et Christophe Lerot, pour leur amitié, leur bonne humeur et aussi pour m’avoir supporté au quotidien.

Je remercie également Caroline Fayt, Christian Hermans, Martine De Mazière, Jean- Christopher Lambert, José Granville, Pierre Gerard, Quentin Errera, Simon Chabrillat, Franck Daerden, Paul Simon, Jean-François Müller, Trissevgeni Stavrakou, Corinne Vigouroux et Bart Dils qui ont tous, de près ou de loin, contribué à mon travail, par un conseil, une discussion ou un support technique.

Mes remerciements vont aussi à ma famille, mes amis et toutes les personnes de mon entourage qui m’ont soutenu lors de ces cinq années. Je tiens enfin à remercier tendrement ma fille Noémie et Laurence, les deux femmes de ma vie.

Ces recherches ont été financées par la Politique Scientifique Fédérale (BELSPO), l’

Agence Spatiale Européenne (ESA) et l’Organisation Européenne pour l’Exploitation de Satellites Météorologiques (EUMETSAT) via les projets AGACC, Prodex/NOy-Bry, Prodex/SECPEA et les projets TEMIS (ESA DUE) et Ozone SAF (BrO Visiting Scientist).

Nicolas Theys Uccle, Novembre 2009

(6)
(7)

Résumé

Les composés bromés jouent un rôle important dans la stratosphère et la troposphère en tant que catalyseurs de la destruction d’ozone. Bien que l’impact des espèces bromées sur la chimie de la stratosphère soit largement compris, il reste de nombreuses incertitudes en ce qui concerne les sources et les pertes de brome dans l’atmosphère ainsi qu’à propos de la chimie et de l’impact des espèces bromées sur la troposphère.

Le travail présenté dans ce manuscrit porte sur la télédétection du monoxyde de brome (BrO) à partir de spectres du rayonnement solaire diffusé, mesurés dans l’ultraviolet à partir d’instruments au sol de type multi-axis et satellitaires à visée nadir (GOME et GOME-2). La méthode de spectroscopie d’absorption optique différentielle est utilisée pour inverser la colonne intégrée de BrO le long du chemin optique effectif dans l’atmosphère. Nous avons développé de nouveaux algorithmes afin de dériver les colonnes verticales de BrO résolues en leurs contributions stratosphérique et troposphérique. Pour la géométrie d’observation au sol, un moyen a été trouvé pour déterminer indépendamment les colonnes de BrO stratosphérique et troposphérique, et cela à partir de la variation diurne des mesures de BrO. Pour les observations satellitaires, la contribution de la stratosphère à la colonne mesurée de BrO est estimée à partir d’une climatologie innovante de BrO stratosphérique. Cette climatologie est basée sur un modèle de pointe de la chimie et du transport de la stratosphère; par ailleurs, elle tient compte explicitement de l’impact de la dynamique atmosphérique et de la photochimie sur la distribution du BrO dans la stratosphère. La fraction troposphérique de la colonne totale mesurée de BrO est quant à elle dérivée à partir d’une technique résiduelle tenant compte des effets des nuages et de la réflectivité de la surface.

Soutenus par un vaste jeu de données corrélatives, les résultats présentés dans ce travail permettent d’étudier de manière appropriée l’évolution spatiale et temporelle du BrO atmosphérique à l’échelle globale. Ces résultats permettent également de mieux estimer l’importance du BrO dans la couche limite planétaire polaire et dans la troposphère libre ainsi que la contribution des précurseurs bromés à temps de vie court sur le budget de brome dans la stratosphère. Nous présentons également la première détection satellitaire de BrO dans un panache volcanique, à l’aide de mesures effectuées avec l’instrument GOME-2.

(8)
(9)

Abstract

Bromine compounds play an important role as catalyst of the ozone destruction in both the stratosphere and troposphere. While the impact of bromine species on the chemistry of the stratosphere is known to a large extend, a number of uncertainties remain regarding the sources and sinks of atmospheric bromine as well as the chemistry and impact of bromine species on the troposphere.

This work describes remote-sensing observations of bromine monoxide (BrO) derived from scattered sunlight spectra in the ultraviolet region measured by ground-based multi- axis and satellite nadir viewing instruments (GOME and GOME-2). The method of differential optical absorption spectroscopy is used to retrieve the columnar concentration of BrO along the effective light path through the atmosphere. New algorithms to derive vertical columns of BrO resolved into their stratospheric and tropospheric contributions are developed and described. For the ground-based geometry a way was found to determine independently the stratospheric and tropospheric BrO columns from the diurnal variation of the BrO measurements. For the satellite observations, the contribution of the stratospheric BrO to the measured column is estimated using an innovative stratospheric BrO climatology. This climatology is based on a state-of-the-art stratospheric chemical transport model, and explicitly accounts for the impact of atmospheric dynamics and photochemistry on the stratospheric BrO distribution. As for the tropospheric fraction of the measured total BrO column, it is derived using a residual technique accounting for the effects of clouds and surface reflectivity.

Supported by an extensive set of correlative data, the results presented here allow to study properly the spatial and temporal evolution of atmospheric BrO at the global scale and enable to better assess the significance of BrO in the polar planetary boundary layer and free-troposphere as well as the contribution from very short-lived brominated sources gases to the stratospheric bromine budget. We also report on the first satellite detection of BrO in a volcanic plume, using GOME-2 measurements.

(10)
(11)

Atmospheric Bromine Monoxide:

multi-platform observations and model calculations

Table of contents

Chapter 1 – Introduction 1 Chapter 2 - General aspects 7

2.1 Dynamics of the atmosphere 7

2.1.1 Vertical structure of the atmosphere 7

2.1.2 Dynamics of the atmosphere influencing trace gases distribution 8

2.2 Stratospheric photochemistry 12

2.2.1 Ozone chemistry 12

2.2.2 Hydrogen and nitrogen chemistry 15

2.2.3 Halogen chemistry and source gases 18

2.2.3.1 Chlorine and bromine source gases 18

2.2.3.2 Chlorine and bromine chemistry 22

2.2.4 Heterogeneous chemistry – the stratospheric ozone hole 26

2.3 Tropospheric bromine chemistry 30

2.3.1 Tropospheric ozone 30

2.3.2 Polar boundary layer bromine 32

2.3.3 Extra-polar boundary layer bromine 36

2.3.4 Free-tropospheric BrO 38

2.4 Open questions related to atmospheric bromine 38

Chapter 3 - Observation principles 41

3.1 Solar spectrum 41

3.2 Interaction processes of radiation in the atmosphere 42

3.2.1 Absorption 43

3.2.2 Elastic scattering 43

3.2.3 Raman scattering 44

3.2.4 Interaction on the surface 45

3.3 Differential Optical Absorption Spectroscopy 46

3.3.1 DOAS principle 47

3.3.2 Particular aspects of the DOAS technique 49

3.3.3 Application: BrO retrieval 52

3.4 Atmospheric radiative transfer 54

(12)

3.4.1 Radiative transfer equation 54

3.4.2 Air Mass Factor calculation 57

Chapter 4 – Instruments 61

4.1 Ground-based measurements 61

4.1.1 Instruments 62

4.1.2 Viewing geometries 63

4.2 Satellite measurements 66

4.2.1 Instruments 66

4.2.2 Viewing geometries, spatial resolution and coverage 69

Chapter 5 – Ground-based multi-axis DOAS BrO observations at

Reunion-Island 73

5.1 Introduction 73

5.2 Data analysis 74

5.2.1 DOAS slant column retrieval 74

5.2.2 Inversion of stratospheric and tropospheric columns 76

5.2.3 Averaging kernels 79

5.2.4 Error analysis 81

5.3 Results and discussion 84

5.3.1 Determination of the aerosol settings 84

5.3.2 Determination of the residual slant column density 86

5.3.3 Clear-sky results 87

5.3.3.1 Tropospheric BrO 88

5.3.3.2 Stratospheric BrO 89

5.3.4 Determination of the tropospheric BrO vertical distribution 91

5.3.5 Seasonal variation 93

5.3.6 Comparison with SCIAMACHY total column BrO observations 94

5.4 Conclusions 95

Chapter 6 – A global stratospheric BrO climatology based

on the BASCOE chemical transport model 97

6.1 Motivation 97

6.2 Model 98

6.2.1 Bromine species 99

6.2.2 Stratospheric aerosol settings 100

6.3 Verification of model results 101

6.3.1 Comparison of modeled and measured stratospheric O3 and NO2 columns 101 6.3.2 Comparison of modeled and measured stratospheric BrO 103

6.3.2.1 Comparison to ground-based stratospheric BrO data 104

6.3.2.2 Comparison with LPMA/DOAS balloon profiles 106

6.3.2.3 Comparison to SCIAMACHY limb profiles 107

6.3.2.4 Discussion 109

(13)

6.4 Stratospheric BrO climatology 111

6.4.1 General approach 111

6.4.2 Dynamics of the stratosphere 112

6.4.3 Bromine monoxide photochemistry 115

6.4.4 Long-term trend in stratospheric bromine 117

6.4.5 Results and error analysis 117

6.5 Conclusions 119

Chapter 7 – Satellite BrO observations 121

7.1 Introduction 121

7.2 Retrieval of tropospheric BrO columns from satellite observations –

Algorithm description and error analysis 122

7.2.1 General description 122

7.2.2 DOAS total slant column retrieval 123

7.2.3 Weighting functions calculation 124

7.2.4 Stratospheric correction 127

7.2.5 Tropospheric air mass factor 127

7.2.6 Error analysis 132

7.3 Retrieval of tropospheric BrO columns from GOME observations on

ERS-2 134

7.3.1 GOME slant column retrieval 134

7.3.1.1 Data analysis 134

7.3.1.2 Results 137

7.3.2 Global tropospheric BrO distribution 140

7.3.3 Latitudinal and seasonal variations of tropospheric BrO 144 7.4 Retrieval of tropospheric BrO columns from GOME-2 observations on

MetOp-A 147

7.4.1 GOME-2 slant column retrieval 147

7.4.2 Transport of tropospheric BrO: case studies 148

7.5 Verification of the retrievals 154

7.5.1 Comparison to ground-based observations 154

7.5.2 Comparison to tropospheric CTM calculations 156

7.6 Conclusions 157

Chapter 8 – Satellite detection of volcanic BrO emissions 159

8.1 Introduction 159

8.2 The Kasatochi volcano 160

8.3 Methods 161

8.3.1 Data analysis 161

8.3.2 Atmospheric transport modeling 161

8.4 Results 162

8.5 Discussion 165

8.6 Conclusions 166

(14)

Chapter 9 – Conclusions and perspectives 167

Appendix A – Inverse problem: The Optimal Estimation Method 171

Appendix B – Slant column density DOAS retrieval: Error analysis 177 Appendix C – List of data sets 181

Publications 183

References 185

List of Acronyms 197

(15)
(16)
(17)

Chapter 1 Introduction

Centered at around 20-25 km altitude, the stratospheric ozone layer plays a major role in protecting life on Earth. Owing to the strong ultraviolet (UV) absorption of ozone in the Hartley-Huggins bands, the incoming solar UV radiation is largely absorbed and converted into heat, affecting temperatures and dynamical flow patterns. Ozone has also strong infrared absorption bands and thus affects the radiation back from Earth into space. This alters the temperature structure of the troposphere and influences the climate.

Ozone is also present in the lower troposphere, where it has important effects on human health (particularly on the respiratory system). In contrast to stratospheric ozone that is of natural origin, tropospheric ozone is mainly a product of pollution.

Since the 1970’s and the pioneering studies of Crutzen (1970) and Molina and Rowland (1974), anthropogenic emissions of nitrogen and halogen compounds have been shown to exert a large influence on the stratospheric ozone budget. This was spectacularly confirmed in 1985 by the discovery of the ozone hole phenomenon over Antarctica (Farman et al., 1985) whereby each spring the ozone layer is almost completely destroyed between 12 and 20 km of altitude. This destruction extends over several tens of million square km. Observations in the northern hemisphere have shown that analogous ozone depletion - although less severe – also occurs in the Arctic regions. The formation of the ozone hole was soon identified to be due to the presence of inorganic chlorine and bromine (hereinafter referred to as Cly and Bry)1 in the stratosphere. A prerequisite is the formation of a strong winterly cyclone over the pole - the polar vortex. Inside the polar vortex, the temperature can drop to very low values, so that polar stratospheric clouds (PSCs) can form. Heterogeneous reactions on the surface of PSCs then convert inorganic halogen compounds such as ClONO2, HCl and BrONO2 into more reactive species as Cl2, HOCl and BrCl (Solomon et al., 1986). When the sunlight returns after polar night, these species are photodissociated and destroy stratospheric ozone through catalytic cycles. In addition to the stratospheric ozone depletion over the poles, a significant although lower ozone loss also occurs at mid-latitudes. The increased levels of chlorine and bromine also contribute to mid-latitudes ozone loss through heterogeneous reactions on the surface of stratospheric background aerosols. It should be noted that bromine species are less abundant by about two orders of magnitude compared to chlorine.

However, on a molecule to molecule basis the ozone depletion potential of bromine is much larger than that of chlorine. Consequently, bromine species account for about 25%

of the ozone loss at mid-latitudes while in polar regions this contribution may reach 50%.

The sources of inorganic halogens in the stratosphere originate from the emissions of organic halogen compounds (mainly man-made chlorofluorocarbons (CFCs), halons, methylchloride and methylbromide) that are transported into the stratosphere before being converted into inorganic forms by photolysis or reaction with O(1D) and OH

1 Xy ≡X+XO+HX+HOX+XONO2+... (X=Cl or Br) is the total number of inorganic molecules.

(18)

radicals. After the recognition of the involvement of halogen species in the depletion of stratospheric ozone, the Montreal Protocol on Substances that Deplete the Ozone Layer was agreed in 1987, leading to the progressive phase-out of many long-lived ozone depleting substances. After five successive amendments to the Montreal Protocol, the cumulative levels of chlorine and bromine from the ozone depleting substances are now decreasing in the atmosphere. Of relevance for the study of the evolution of the ozone layer is the distinction between the source gases that are regulated by the Montreal Protocol and the naturally emitted halogen-containing compounds. In that respect, bromine compounds are of particular interest because about half of the stratospheric Bry

loading is due to natural sources (compared to about 15% for chlorine). These source gases are generally released by biological processes (e.g., in the oceans) and are characterized by low atmospheric lifetimes (typically <6 months).

After the role of bromine in the destruction of stratospheric ozone was highlighted, it became soon clear that inorganic bromine compounds could also have a significant effect in the troposphere (for an overview see, e.g., von Glasow and Crutzen, 2007). In the boundary layer, it was found that large amounts of inorganic bromine are seasonally released in polar regions in both hemispheres during spring due to a phenomenon known as Polar-bromine explosion. These emissions are responsible of complete ozone depletion events in the polar boundary layer, as well as interactions with mercury chemistry.

Although more localized, inorganic bromine emissions have also been identified over salt lakes, in the marine boundary layer and in volcanic plumes. Furthermore, observations from space, the ground and balloons have shown that inorganic bromine may be produced and sustained in the free troposphere at the global scale. Elaborating on these observations, modeling results have shown that free-tropospheric bromine might have a significant impact on tropospheric ozone (and on tropospheric chemistry in general), leading to a reduction in the O3 mixing ratio of up to 40% locally.

Currently, several scientific issues regarding atmospheric bromine still remain:

- The sources and sinks of the bromine short-lived species, as well as their regional variability, are characterized by large uncertainties. Much research is devoted to assess the importance of these organic bromine short-lived species on the stratospheric chemistry. How the very short-lived substances enter the stratosphere through the tropical tropopause is also not fully characterized (World Meteorological Organization report 2007, Chapter 2 on halogenated very short-lived substances).

Furthermore, many of the underlying physical and chemical processes leading to the release of inorganic bromine to the stratosphere are potentially sensitive to climate change (Salawitch, 2006).

- The exact mechanism, which leads to an initial bromine release in the polar boundary layer as well as the influence of transport and chemical processes on bromine, is still not clearly characterized (Simpson et al., 2007).

- The origin of free-tropospheric bromine is not fully understood. It has been speculated that it might be due to uplifting of surface Bry, downward transport from the stratosphere, volcanic activity or the decomposition of short-lived organic bromine compounds under the action of heterogeneous and/or gas-phase photochemical reactions.

(19)

The active form of bromine, taking part in the destruction of ozone, is bromine monoxide (BrO). During daytime, BrO is the most abundant inorganic bromine species (typically 30 to 70% of total Bry). It is also the bromine species that can be measured with highest accuracy using absorption spectroscopy techniques1. Over the last two decades, many observations of BrO were performed using instruments based on ground, balloon, aircraft and satellite platforms. These measurements have contributed significantly to improve our understanding of atmospheric bromine. To a large extent, they were achieved through the exploitation of BrO electronic bands in the near ultraviolet2 and using the technique of differential optical absorption spectroscopy (DOAS; Platt and Stutz, 2008).

The objective of this PhD thesis is to contribute to a better understanding of atmospheric bromine by studying the BrO content in both the troposphere and stratosphere and by characterizing quantitatively its spatial and seasonal variations. For this purpose, we will determine vertically integrated concentrations (vertical column amount) of BrO in both the troposphere and stratosphere by using (1) ground-based and satellite nadir DOAS instruments, and (2) stratospheric model calculations.

This thesis is organized in nine chapters. Chapter 2 gives an overview of the role of bromine and its chemistry in the atmosphere. The observation principles and the DOAS instruments are described in chapters 3 and 4 respectively. Ground-based and satellite instruments used in this work have in common that they all measure the solar UV radiation scattered from the atmosphere and reflected by the ground; the analysis of the measured spectra with the DOAS method then provides the BrO absorption averaged over all light paths contributing to the signal. The retrieved quantity can be regarded as the integrated BrO concentration along the mean optical light path (it is often referred to as the ‘slant column density’). Since the light that reaches the instrument travels through the entire atmosphere, the measured slant column contains absorption from both stratosphere and troposphere. Evaluating the tropospheric and stratospheric BrO columns from the measurements requires essentially two additional steps which are at the heart of our study:

- the separation of the stratospheric and tropospheric contributions to the measured BrO slant columns.

- the conversion of the (tropospheric and stratospheric) slant columns into vertical columns, using adequate modeling of the transfer of the radiation in the atmosphere.

Important parameters affecting the optical light path (such as the observation geometry, surface reflectivity, atmospheric gases, clouds and aerosols) need to be accounted for.

In practice, the inversion of stratospheric and tropospheric BrO columns from an individual DOAS observation is an ill-posed problem. In chapter 5, we present a retrieval method that enables the inversion of independent tropospheric and stratospheric BrO columns from a combination of several observations made from a ground-based

1 Besides BrO, only few measurements of inorganic bromine species exist: HBr, HOBr and recently stratospheric BrONO2 (Höpfner et al., 2009). Note that it is also possible to detect several organic and inorganic bromine compounds using in-situ measurements.

2 Rotational lines in the microwave region can also be used to detect BrO (Kovalenko et al., 2007) but the estimated precision and accuracy is less good.

(20)

platform. In contrast to satellite nadir instruments that provide snapshots in time1, ground-based instruments have the great advantage to measure the atmosphere at a given location continuously during the day. Moreover, with the ground-based system used here, the same air mass is sounded sequentially under different viewing directions, from the horizon to the zenith (Multi-axis DOAS or MAXDOAS measurements). It is therefore particularly well suited to combine the measurements acquired from noon to twilight in the different viewing directions to infer the stratospheric and tropospheric BrO columns.

This stratosphere-troposphere separation is achieved by exploiting the different evolution of the measurement sensitivity in the stratosphere and the troposphere as a function of solar zenith angle and viewing directions. We have applied this retrieval algorithm to measurements performed at Reunion Island (21.06°S, 55.47°E, Indian Ocean). The location of this site is specially interesting since only few BrO measurements have been performed in tropical region so far, in spite of the importance of the tropics for the atmosphere (and besides for setting the chemical boundary conditions of the stratosphere). In particular, a remote area as Reunion-Island is ideal to investigate the delivery of Bry from oceanic short-lived bromine species to the free-troposphere and lower stratosphere. Moreover, tropical BrO measurements are also needed to validate (and even constrain) satellite data. Indeed, the measurement conditions at low latitudes (high sun and short light path through the atmosphere) make the BrO retrieval particularly challenging for the satellite instrument.

The method developed for the ground-based observations to separate the tropospheric and stratospheric BrO columns, can not be transposed to satellite nadir observations. The strategy we adopted in this thesis is, in a first step, to build a consolidated stratospheric BrO dataset that can be used afterwards as reference to separate with good accuracy the stratospheric and tropospheric contributions to the satellite measured BrO columns. This is the subject of chapter 6. We have developed a new method to determine stratospheric BrO concentration profiles (and columns) using a climatological approach. The proposed climatology is able to reproduce the important spatial and temporal variations of stratospheric BrO by using dynamical and chemical indicators. The adopted parameterization is based on output data from a global 3-D model of the chemistry and transport of the stratosphere (BASCOE). The model simulations are optimized for bromine species by the implementation of an up-to-date photochemistry and a realistic total bromine budget (including a contribution from short-lived bromine source gases).

We have made comparisons of modeled BrO and an extensive data set of stratospheric BrO observations from ground-based (including the reference set of Reunion Island), balloon and satellite limb-viewing instruments. There are three reasons for this validation exercise: (1) test our understanding of stratospheric bromine, (2) assess the consistency of the stratospheric BrO observational data set, and (3) extend the observations at the global scale via the model.

Chapter 7 deals with the determination of global stratospheric and tropospheric BrO columns using GOME and GOME-2 satellite nadir observations. The stratospheric BrO columns are estimated using the stratospheric BrO climatology (underlying the currently available stratospheric BrO observations). Although based on model simulations, the calculated stratospheric BrO columns are still representative of the sounded air masses as they are evaluated based on measured quantities. The tropospheric BrO columns are

1 This is true for the low Earth orbit satellites used here, but obviously not for geostationary orbit platforms.

(21)

estimated from a residual technique that combines measured and calculated stratospheric columns, and that accounts for the impact of clouds, surface reflectivity and viewing geometry on the measurement sensitivity. For the first time, it was possible to properly quantify the BrO amount in the global troposphere using satellite nadir instruments. The seasonal variation and spatial distribution of the retrieved tropospheric BrO columns are interpreted in the light of our current understanding of tropospheric bromine. In particular, the existence of a free-tropospheric BrO background at the global scale is investigated. It is also shown that our method enables to separate the large-scale stratospheric BrO structures from those of tropospheric origin in the total BrO column field measured from space. It allows studying unambiguously the development and transport of BrO plumes in the polar boundary layer. The tropospheric BrO results were compared to correlative ground-based data as well as modeling calculations and a good agreement was found.

Chapter 8 presents a side study focused on the satellite detection of volcanic BrO emissions after the major eruption of the Kasatochi volcano in Alaska (52.17°N, 175.51°W) on August 2008. Using a Lagrangian trajectory model, it was possible to simulate the transport of the volcanic plume and estimate the total mass of reactive bromine emitted by the volcano. These results represent the first ever space-based observation of BrO released by volcanic activity. This study further strengthens the idea that volcanoes probably play an important role as a source of bromine in the atmosphere.

In chapter 9, this study ends with the conclusions and perspectives.

(22)
(23)

Chapter 2

General aspects

In this chapter some aspects concerning atmospheric dynamics and chemistry will be introduced. A strong emphasis is given on the role of the bromine species in the atmosphere, and on the physical and chemical processes influencing their distribution in the stratosphere and the troposphere.

2.1 Dynamics of the atmosphere

2.1.1 Vertical structure of the atmosphere

The atmosphere (from Greek ατμός - atmos, “vapor” + σφαίρα - sphaira, “sphere”) of the Earth is commonly described as a series of layers or “spheres” defined by their thermal characteristics (see Figure 2.1).

Figure 2.1 Vertical structure of the Earth’s atmosphere. The temperature profile is taken from the U. S. Standard Atmosphere (Anderson et al., 1986).

(24)

Specifically, each layer is a region where the change in temperature with respect to the altitude (lapse rate) has a constant sign. The layer above the surface is known as the troposphere, and contains approximately 75% of the atmosphere’s mass. The troposphere is characterized by a decrease of the temperature with altitude (at a mean lapse rate of 6.5 K/km), due to the adiabatic expansion of rising air and the consequent cooling. The troposphere is therefore characterized by convection and strong mixing of air masses. It is the layer where most of the weather phenomena and atmospheric turbulence occur. The troposphere is usually divided in two regions: (1) the boundary layer, directly influenced by the presence of the Earth’s surface, responds to surface forcing with a time scale of about an hour or less. The convective air motions generate intense turbulent mixing. The boundary layer has a thickness varying from 300 m to 2000 m, depending on the local atmospheric energy budget. (2) the free-troposphere is the layer above the boundary layer, which is not directly affected by the surface. Long-range transport and vertical motions of gases and fine particles (aerosols) are frequent features in that region. The upper boundary of the troposphere is called the tropopause and is marked by a sharp change of the lapse rate1. The tropopause temperature and height depends on the latitude, season and daily changes in meteorological conditions. Typical tropopause height for tropical regions is of 16 to 18 km, and the corresponding temperature is about 200K, while in the polar regions the tropopause elevation is only of about 8 km, and the temperature roughly 220 K.

Above the tropopause, temperature first remains constant and then increases in the stratosphere, due to ozone heating by absorption of ultraviolet radiation from the sun. The vertical stratification, with warmer layers above and cooler layers below, makes the stratosphere dynamically stable. In contrast to the troposphere, the stratosphere involves only weak vertical motions and is dominated by radiative processes. The stratosphere extends to about 50 km, where the temperature exhibits a maximum at the stratopause. At higher altitude, the temperature again decreases up to 90 km, where another temperature minimum is found. This layer is called the mesosphere and its upper boundary is the mesopause. The region above the mesopause is called the thermosphere and is characterized by an increase of temperature with height, due to the absorption of extreme ultraviolet radiation by molecular and atomic oxygen.

The work presented in this thesis is concerned by aspects related to the so-called middle atmosphere, i.e., the troposphere and the stratosphere only.

2.1.2 Dynamics of the atmosphere influencing trace gases distribution

The distributions of the atmospheric constituents result from the competition of atmospheric dynamics and chemistry. The relative importance of both effects can be evaluated from an analysis of the time constants related to these dynamical and chemical processes (production and destruction). While several species are only influenced by photochemistry, other atmospheric constituents are characterized by time constants related to chemical processes that are much larger than the time constants associated to the dynamics in the atmosphere. In this case, the distributions of these constituent are

1 The thermal tropopause height is defined by the World Meteorological Organization (WMO) as the height of the base of a layer at least 2km thick, in which the rate of decrease of temperature with height is less than 2 K/km.

(25)

solely influenced by dynamical effects and the chemical constituent can be considered as a good indicator of the dynamic of the atmosphere1. This section aims presenting some basis on the large scale atmospheric circulation patterns which contribute to the spatial distribution of trace gases2. A general description of the transport processes in the troposphere and the stratosphere will be given. The reader is referred to the literature (e.g., M. Salby; Holton et al., 1995) for a detailed treatment of the theory of the atmospheric transport.

Tropospheric circulation

Unequal heating is the main driving mechanism responsible for the global tropospheric circulation. At the equator, warm moist air is lifted aloft in low pressure areas. The cooling of rising air causes the formation of clouds by condensation of water vapor. The temperature inversion at the tropopause then acts as a ‘roof’ and hinders the upward motion of air, causing it to spread and move poleward. At about 30° latitude, dry air descends in a high pressure area. Some of the air masses travel back to the equatorial region, closing the loop of the convection cell. Such large scale convection systems are observed in both hemispheres and are known as the Hadley cells (see Figure 2.2).

Figure 2.2 Schematic view of the global circulation in the troposphere.

© http://www.sbg.ac.at/ipk/avstudio/pierofun/atmo/el-scans/hadley.jpg

1 Hereafter we will refer to the widely used term ‘tracer of the atmospheric dynamics’. Note that it is possible to demonstrate using the law of conservation of mass that the volume mixing ratio of a tracer is conserved during atmospheric transport.

2 The term ‘trace gas’ refers to a gas which makes up less than 1% by volume of the earth's atmosphere. It includes all gases except nitrogen (78.1%) and oxygen (20.9%).

(26)

In addition, the air flow is also deflected from west to east under the action of the Coriolis force, leading to the formation of the well-known near-surface trade winds and the subtropical jet winds (typically located between 10 and 16 km).

In the polar regions, air masses at ~60° latitude are sufficiently warm and moist to undergo convection as well and drive a thermal loop (called the polar cell). Air circulates within the troposphere, limited vertically by the tropopause at about 8 km. Warm air rises at lower latitudes and moves poleward through the upper troposphere at both the North and South poles. When the air reaches the polar areas, it has cooled considerably, and descends as a cold, dry high pressure area, moving away from the pole along the surface, but twisting westward as a result of the strong Coriolis effect to produce the Polar easterlies.

A third circulation cell, the Ferrel cell, is also present between ~ 30° and 60° latitudes (in both hemispheres) and results mainly from the entrainment of air masses by the Hadley and polar cells. The wind along the surface in the Ferrel cell is westerly in both hemispheres due to the Coriolis force. Noteworthy is the presence of fast flowing airs within the transitional zone between the tropopause and the Ferrel cell (polar jet streams).

Stratospheric circulation

The transport in the stratosphere is fundamentally different from what is occurring in the troposphere. Indeed, the temperature inversion above the tropopause effectively inhibits vertical transport in the stratosphere, and the important convection cells and subsequent mixing of air typical of the troposphere are not observed in the stratosphere.

The dominant feature of the global stratospheric wind system is the presence of strong mean zonal winds (wind in the longitudinal direction), which tend to eliminate the longitudinal gradients of the chemical constituents in the stratosphere. As a first approximation, it can be assumed that the air masses are transported adiabatically (motion along quasi-horizontal isentropic surfaces). Solving the fundamental equations of the fluid mechanics for adiabatic parcel displacements leads to the conceptual view that zonal winds are thermal winds, in the sense that the zonal wind field is proportional to the horizontal temperature gradient. As a consequence of the temperature difference between the tropical lower stratosphere and the polar stratosphere, an easterly wind forms in the summer hemisphere, whereas a westerly wind appears in the winter hemisphere.

The most striking dynamical structure observed in the stratosphere, related to the presence of zonal winds, is probably the polar vortex. During the polar winter night, the stratosphere is not exposed to sunlight and the air masses undergo a severe cooling. A large temperature gradient appears between the mid-latitudes and the pole, and hence leads to the formation of a rapid circumpolar wind. The strong jet wind at the edge of the so-called vortex acts a dynamical barrier and has the effect to isolate the air over the polar region. It should be noted that some diabatic processes have an important influence on the stability of the polar vortex, and are responsible at the end of its dislocation by mixing of air. As a matter of fact, at the beginning of spring, the sun is heating the polar stratospheric air masses again, and weakens the polar vortex. In addition, the polar vortex is further disturbed and destabilized by atmospheric waves (called planetary waves or Rossby waves), forced from beneath and propagating into the stratosphere. These waves result of the different heat capacity of land and water and are induced by the build-up of

(27)

temperature and pressure gradients. Their amplitudes increase over a more variable surface topography. Whereas the Antarctic continent is centered on the pole and surrounded solely by water, the topography of the northern hemisphere is more complex.

This leads to a more stable and circular vortex in the southern hemisphere than in the northern hemisphere. We will see in the next section that it has a crucial impact on the depletion of polar stratospheric ozone. The polar vortex is not the unique example of dynamical barrier in the stratosphere. Indeed, the difference in temperature between the tropical lower stratosphere and the extra-tropical stratosphere is responsible for the appearance of the subtropical jet winds located around 30°N and 30°S. These strong winds form the so-called tropical barrier which has the effect to isolate the tropics from the extra-tropical air masses.

Besides the zonal circulation described above, meridional (in the latitudinal direction) and vertical transport of air masses plays also a major role on the distribution of chemical species in the stratosphere. Note however that the mean meridional and vertical winds in the stratosphere are very small compared to those of the zonal winds. The overall feature of the mean stratospheric meridional-vertical circulation is well described by the Brewer- Dobson circulation, and is illustrated in Figure 2.3.

Figure 2.3 Schematic diagram showing the Brewer-Dobson stratospheric circulation.

The tropopause is shown by the thick line. Thin lines are isentropic or constant potential temperature1 surfaces labeled in Kelvins. The shaded regions represent the dynamical barriers in the stratosphere (polar vortex and tropical barriers). © NASA. Studying Earth’s Environment from Space (http://www.ccpo.odu.edu/SEES/index.html).

1 The potential temperature is a quantity conserved in an adiabatic air motion. It represents the temperature which an air parcel would attain if it were adiabatically compressed or expanded starting from a temperature T and pressure p to a pressure of 1000 mb, and is given by: θ=T(1000/p)κ where κ=R/cp=0.286.

(28)

Recognition of such stratospheric transport is historically based on observations of long- lived chemical species (water vapor and ozone) that suggested a circulation exhibiting a rising motion only in the tropics, and descending motion at extra-tropical latitudes. The Brewer-Dobson circulation brings air masses from the tropics to mid- and high-latitudes.

In the winter hemisphere, a large-scale descent of air masses from the upper stratosphere is observed in the polar vortex1. Transport of air masses is also arising from the tropical region toward the pole through the ‘extra-tropical pump’, which is largely driven by the breaking of Rossby waves in the so-called ‘surf zone’ of mid-latitudes. It should be emphasized that the Brewer-Dobson circulation is a very slow process, so that the air will cycle through the stratosphere within ~ 5 years.

Despite the fact that the tropopause tends to isolate the stratosphere from the troposphere (because of the temperature gradient inversion), several pathways are possible for stratosphere-troposphere exchange of mass and chemical species. Clearly, the strong tropical convection is responsible for the injection of constituents in the stratosphere (often because of the fast transport of air due to the formation of tropical convective clouds). The large-scale descent of air in the polar region brings air back to the troposphere. According to Figure 2.3, stratosphere-troposphere exchange is also possible at mid-latitudes, through adiabatic transport (some isentropic surfaces cross the tropopause) or wave-induced forcing.

2.2 Stratospheric photochemistry

In this section, photochemical processes relevant to important constituents of the stratosphere will be discussed. We first focus on the photochemistry of ozone, hydrogen, nitrogen and halogen compounds. We will then describe the physical and chemical processes taking place in the stratosphere leading to the so-called stratospheric ozone hole.

2.2.1 Ozone chemistry

One of the most important stratospheric species is ozone (O3). Although ozone is present in both stratosphere and troposphere, the bulk of ozone resides in the stratosphere (with the maximum concentration around ~22 km) and constitutes the commonly referred ozone layer. Despite the small concentration of ozone in the stratosphere, it is playing an essential role for the protection of living organisms on Earth2 by absorbing most of the harmful ultraviolet (UV) radiation emitted from the sun. The absorption of short wavelength solar radiation by ozone leads also to the heating of the stratosphere, which in turn drives the stratospheric circulation. Furthermore, ozone interacts with long wavelength radiation (infrared radiation) and hence contributes to the greenhouse effect of the Earth’s atmosphere.

A photochemical scheme for the formation and destruction of ozone in the stratosphere based on oxygen-only chemistry was first proposed by Chapman (1930):

1 This phenomenon is sometimes called ‘air subsidence’.

2 The progressive formation of the ozone layer has constituted a fundamental prerequisite for the evolution of life outside the oceans, about 400 million years ago.

(29)

O2 + hν → O + O (λ<242 nm) (R 1) O + O2 + M → O3 + M (R 2) O3 + hν → O(1D) + O2 (λ <308 nm) (R 3) O(1D) + M → O(3P) + M (R 4) O3 + hν → O(3P) + O2 (λ <1180 nm) (R 5) O + O + M → O2 + M (R 6)

O(3P) + O3 → 2 O2 (R 7)

where M is any arbitrary molecule. The fast interchange between ozone and atomic oxygen through the Chapman cycle leads to a photochemical equilibrium explaining the persistence of a stratospheric ozone layer. According to this simple model of ozone production, the photodissociation of O2 is maximum with overhead sun. Ozone is thus mostly produced in the equatorial region (even if the production rate is still effective at higher latitudes). Moreover, the production of ozone occurs mainly in the upper stratosphere, because of the strong attenuation of the short wavelength radiation that controls the photodissociation of O2 (R 1).

It should be noted that the individual species of the Chapman cycle have a short lifetime (e.g., about an half hour for O3), but if we define the OX family as the sum of the odd- oxygen species (OX=O3+O(1D)+O(3P)) then the reactions (R 1-7) only convert one species of the OX family into another and the lifetime of OX as a whole is much longer, about several months to a year in the lower stratosphere (at the altitude of the maximum ozone concentration). Odd-oxygen species can thus be transported from the equatorial region toward the poles1, as a result of the Brewer-Dobson circulation. Hence, the global distribution of ozone depends on latitude and season. This is illustrated in Figure 2.4 that shows the average amount of total ozone column2 at a given latitude band as a function of time, as recorded by the TOMS satellite instrument between 1979 and 1992. One can see that the ozone column is rather small at the equator and increases toward the poles.

This is due to the peak in ozone mixing ratio occurring in the polar regions at lower altitudes, where the increased pressure corresponds to increased molecule numbers. It can also be seen that the maximum of ozone column is found at northern high-latitudes in spring. Note that this maximum is surprisingly not found at high-latitudes in austral spring, as it would have been expected from the hemispheric alternation of the Brewer- Dobson circulation. Instead, a dramatic decrease of the ozone column is observed – the stratospheric ‘ozone hole’.

Well before the discovery of the ozone hole by Farman et al. (1985), it was recognized that the mechanism of Chapman was not sufficient to explain the observations of ozone.

1 In the stratosphere ozone is much more abundant than atomic oxygen so that OX ≈ O3. Hence ozone can be considered to a certain point as a tracer of the dynamics of the stratosphere.

2 The thickness of the ozone layer is determined by the amount of ozone molecules in a column overhead per surface unit, and is calculated by integrating the ozone concentration profile along the vertical axis. The ozone column is often expressed in Dobson units (DU) – one DU refers to a layer of ozone that would be 10 µm thick under standard temperature and pressure.

(30)

Figure 2.4 Mean annual cycle of the total ozone column (expressed in DU) as measured by the TOMS satellite instrument (1979-1992). © NASA. Studying Earth’s Environment from Space (http://www.ccpo.odu.edu/SEES/index.html).

Indeed, the ozone concentrations modeled by the Chapman cycle are largely overestimating the observations of ozone. In particular, the oxygen-only chemical scheme predicts a maximum of the ozone column at the equator which is in total contradiction with the observations. It became clear that stratospheric ozone was not only chemically destroyed by photolysis or reaction by atomic oxygen but also by catalytic reactions involving hydrogen, nitrogen, chlorine and bromine species:

X + O3 → XO + O2 (R 8)

XO + O → X + O2 (R 9)

net: O3 + O → 2 O2 (R 10) where X represents one of the radicals OH, NO, Cl or Br. The radical cycle is regenerated by reaction (R 9) and is available for another ozone destruction cycle until it is removed by a sink process. Thus a small quantity of X can have a large impact on the ozone concentration.

Another catalytic cycle involving species of different families is:

X + O3 → XO + O2 (R 11)

Y + O3 → YO + O2 (R 12)

XO + YO → X + Y + O2 (R 13) net: 2 O3 → 3 O2 (R 14) where X=OH and Y=Cl, X=OH and Y=Br or X=Cl and Y=Br. Since no atomic oxygen is consumed, this cycle is particularly important in the lower stratosphere where the recycling of the radicals by reaction with O (R 9) is less effective.

(31)

The relative contributions of the various cycles to the ozone depletion vary as a function of altitude, latitude, season and local photochemical conditions. Furthermore, the abundance of the ozone-depleted substances is related to the source gases that are closely linked to the biogenic and anthropogenic activities at the surface of the Earth. Most of the source gases are long-lived species, so they can be transported to the stratosphere and be converted into radicals through photolysis or oxidation (by OH or O(1D)).

The following sections are dedicated to the study of the hydrogen, nitrogen and halogen compounds, with respect to the photochemistry and the source gases. The main concepts related to ozone depletion are well documented (see e.g., Solomon, 1999), and here we intend to give a brief description of the key processes.

2.2.2 Hydrogen and nitrogen chemistry

Hydrogen chemistry

The involvement of hydrogen in catalytic ozone depletion cycles was first proposed by Bates and Nicolet (1950). The reactive hydrogen species of H, OH, HO2 and H2O2 are often denoted by the term HOx. The catalytic ozone destruction cycles involving HOx

dominate the ozone loss in the lowermost stratosphere (below 20 km):

OH + O → HO2 (R 15) HO2 + O → OH + O2 (R 16) OH + O3 → HO2 + O2 (R 17) HO2 + O3 → OH + 2 O2 (R 18) net: 2 O + 2 O3 → 4 O2

The HOx species are formed in the stratosphere mainly by the (slow) reaction of excited O atoms with H-containing atmospheric species like water vapor (H2O) and methane (CH4):

H2O + O(1D) → 2 OH (R 19) CH4 + O(1D) → CH3 + OH (R 20) H2O + hν → H + OH (R 21) The major sink processes for the hydroxyl radical are the reaction with HO2, nitric (HNO3) and hydrochloric (HCl) acids:

OH + HO2 → H2O + O2 (R 22) OH + HNO3 → H2O + NO3 (R 23) OH + HCl → H2O +Cl (R 24)

(32)

Nitrogen chemistry

The stratospheric ozone loss between 25 and 40 km is predominantly caused by nitrogen catalytic cycles (Crutzen, 1970; Johnston, 1971):

NO + O3 → NO2 + O2 (R 25) NO2 + O → NO + O2 (R 26) net: O3 + O → 2 O2

NO + O3 → NO2 + O2 (R 27) NO2 + O3 → NO3 + O2 (R 28) NO3 + hν → NO + O2 (R 29) net: 2O3 → 3 O2

NO and NO2 are in a photochemical steady-state during daytime via the fast reactions (R 25-29) and the photodissociation of NO2

NO2 + hν → NO + O (R 30)

and form the so-called reactive NOx (=NO+NO2) family1.

The impact of the reactive nitrogen species on the ozone destruction is however moderated by a number of three-body reactions converting NOx species into reservoirs species:

NO2 + NO3 + M → N2O5 + M (R 31) NO2 + OH + M → HNO3 + M (R 32) NO2 + HO2 + M → HO2NO2 + M (R 33) The reactive species can also be released from the reservoirs by:

N2O5 + M → NO3 + NO2 + M (R 34)

N2O5 + hν → NO3 + NO2 (R 35)

HNO3 + hν → OH + NO2 (R 36)

HNO3 + OH → NO3 +H2O (R 37)

HO2NO2 + hν → HO2 + NO2 (R 38)

There also exist several reactions between NOx and reactive halogenated species that form important halogenated-nitrogen reservoirs. We have intentionally omitted to list these reactions here, as they will be treated in the next section which is dedicated to the chemistry of halogen compounds in the stratosphere.

The reactions listed above largely explain the observed diurnal variation of NOx species:

Dinitrogen pentoxide (N2O5) is produced (through R 31) almost entirely at night (because NO3 is rapidly photodissociated during daytime), and thus constitutes an important nighttime nitrogen reservoir. The fast photolysis of NO2 together with the slow photolysis

1 We also define the NOy family as the sum of all nitrogen species (reactive and reservoirs nitrogen species), i.e. NOy= NO+NO2+NO3+2N2O5+HNO3+HO2NO2+ClONO2+BrONO2

(33)

Figure 2.5 Diurnal variation of NO, NO2, N2O5 (X2) concentrations at 30 km height, calculated by the PSCBOX photochemical model (courtesy of F. Hendrick). The data are shown for Observatoire de Haute-Provence (44°N, 6°E) for 15 March 2004. The dashed lines show the time of sunrise and sunset (90° of solar zenith angle).

of N2O5 is responsible for the typical diurnal variation of NO2 (see Figure 2.5). As the sun begins to rise, a strong decrease of NO2 concentration is observed, due to photolysis.

During the day, the NO2 concentration slowly increases because of N2O5 photolysis. At sunset, the concentration of NO2 increases rapidly due to the absence of photodissocation.

The presence of NOy in the stratosphere is primarily due to the decomposition of nitrous oxide (N2O) emitted at the surface:

N2O + O(1D) → 2 NO (R 39) Nitrous oxide is produced by denitrifying and nitrifying soil bacteria. Other sources contributing to the stratospheric NOy budget include NOx transported to the stratosphere and produced by solar proton events and cosmic rays (in the mesosphere and thermosphere) or by tropospheric lighting.

The major removal processes for stratospheric NOx is via the formation of nitric acid (R 32). The atmospheric lifetime of HNO3 is of ~1 month in the lower stratosphere, because of the relative inefficiency of reactions (R 36-37). Nitric acid has also the important property of being soluble, so that HNO3 can be uptaken into aqueous and solid particles.

Another important heterogeneous reaction that favored the conversion of NOy into nitric acid in the liquid or solid phase (phenomenon called denoxification) is the hydrolysis of N2O5:

N2O5 + H2O (s) → 2 HNO3 (s) (R 40) Here, (s) refers to the solid or liquid phase.

(34)

The irreversible removal of stratospheric NOy (denitrification) occurs by the progressive sedimentation in the polar regions of large particles that contain a substantial amount of HNO3.

The heterogeneous reactions implying nitrogen and halogen species are not listed above, but will be treated in section 2.2.4.

2.2.3 Halogen chemistry and source gases

Halogen species (F, Cl, Br, I) play an important role in stratospheric ozone destruction and are responsible for the formation of the polar ozone hole (see section 2.2.4). The halogen atoms released in the stratosphere from the decomposition of halogenated organic source gases can form acids and nitrates (through reaction with NO2). In case of fluorine, the acid HF is rapidly and irreversibly formed, hence fluorine has a negligible impact on ozone. Chlorine forms both HCl and ClONO2 reservoirs. These gases can, contrary to fluorine, be reconverted to chlorine atoms by gas-phase or heterogeneous chemistry (see below). The amount of chlorine available for ozone destruction cycles depends critically on the chemical rates of destruction and formation of the chlorine reservoirs, i.e. on the partitioning of chlorine between the reactive gases and the reservoirs. Bromine is less bonded than chlorine because the reservoirs (mostly BrONO2) photolyse and react rapidly with OH. Bromine is thus more effective for ozone loss than chlorine. Iodine may also participate to the stratospheric destruction, but its contribution to stratospheric ozone loss is believed to be very small because the iodine-containing gases are largely removed in the troposphere before reaching the stratosphere.

Chlorine is responsible of about 60% of the ozone destruction under ozone hole conditions. Despite its low abundance in the stratosphere, bromine contributes to ~ 30%

of polar ozone loss owing to an Ozone Depletion Potential (ODP)1 that is ~ 60 times higher than chlorine.

2.2.3.1 Chlorine and bromine source gases

The various sources of reactive halogens in the stratosphere are halocarbon gases of anthropogenic and natural origin. Once emitted at the Earth’s surface, they all experience similar processes. After accumulating in the troposphere, the halogen source gases are transported to the stratosphere (mainly in the tropics) were they are progressively converted to Cl and Br by photolysis or reactions with OH or O(1D). The atomic Cl and Br then react with O3 (chemical ozone destruction) to form halogen oxides (ClO and BrO) that are later on converted (by a number of reactions described in section 2.2.3.2) into other stratospheric inorganic halogen species2. Source gases with a long lifetime and low solubility in water have a great chance of reaching the stratosphere. In contrast, gases with the shortest lifetime are to a large extent converted into inorganic halogens species

1 The ODP of a compound is a relative measure of its ability to destroy stratospheric ozone. It is calculated on a ‘per mass’ basis, as the ratio of the total amounts of ozone destroyed by the compound and by the same mass of CFC-11 (see below).

2 We define the inorganic halogen families:

inorganic chlorine Cly=Cl+ClO+2Cl2O2+ClOO+OClO+ClONO2+HOCl+HCl+BrCl and inorganic bromine Bry=Br+BrO+BONO2+HOBr+HBr+BrCl.

(35)

already in the troposphere where they are mainly removed from the atmosphere through washout by falling ice or rain. Therefore, only a small fraction of these source gases enter into the stratosphere. The most important sources of chlorine and bromine in the stratosphere are the chlorofluorocarbons (CFCs) and halons (bromine containing species), industrially manufactured since the 1950’s. These compounds have been widely used during the 20th century as refrigerants (freons), foam blowing agents, solvents, aerosol spray propellants, fire extinguishing agents and chemical reagents. The lifetime of these gases is high, varying from years to centuries and resulting in almost uniform distribution in the troposphere. Table 2.1 presents an overview of the chemical lifetime and abundance of the main halogen containing species.

CFCs, along with carbon tetrachloride (CCl4)and methyl chloroform (CH3CCl3) are the most relevant chlorine source gases of anthropogenic origin. Methyl chloride (CH3Cl) is the only important natural chlorine source species. Based on a budget of the different chlorine source gases and their chemical lifetimes, the delivery of Cly in the stratosphere is estimated to be about 3400 parts per trillion volume (pptv).

The major source of bromine is methyl bromide (CH3Br) which provides more than 50%

of the bromine content from long-lived source species. CH3Br is released by natural (biomass burning, oceans) and anthropogenic (e.g., soil fumigation) processes. Additional sources of bromine are the man-made halogenated hydrocarbons gases (mainly halon- 1211 and 1301) initially developed to extinguish fires.

Table 2.1 Atmospheric lifetime and mole fraction (in 2004) for the primary halogen source gases of chlorine and bromine for the stratosphere. Adapted from WMO report 2007.

Common name Chemical formula Lifetime (years)

Mole fraction in 2004 (pptv) Chlorine

CFC-12 CCl2F2 100 540

CFC-11 CCl3F 45 254

CFC-113 CCl2FCClF2 85 79

HCFCs CHxClyFz 1-20 205

Carbon tetrachloride CCl4 26 95

Methyl chloride CH3Cl 1 527

Methyl chloroform CH3CCl3 0.5 22

Bromine

Halon-1211 CBrClF2 16 4.5

Halon-1301 CBrF3 65 2.8

other halons 3-20 ~ 1

Methyl bromide CH3Br 0.7 9

Short lived species 12-150 days variable

The estimate of the stratospheric Bry loading due to long-lived organic bromine compounds is of about 16-17 pptv (Wamsley et al., 1998). However, several recent studies based on measurements of stratospheric BrO using remote-sensing UV-visible

(36)

techniques (see chapters 3 and 4) from ground-based (Sinnhuber et al., 2002; Schofield et al., 2004 and 2006; Hendrick et al., 2007; Theys et al., 2007), balloon-borne (Pundt et al., 2002; Salawitch et al., 2005; Dorf et al., 2006a,b and 2008) and space-borne limb (Sinnhuber et al., 2005; Sioris et al., 2006) instruments have inferred a total inorganic bromine loading of 18-25 pptv, suggesting that an additional contribution must be considered, possibly due to bromine release from short-lived biogenic organic compounds (such as CHBr3, CH2Br2, CH2BrCl, CHBr2Cl, CHBrCl2, CH2BrCH2Br) or even through direct injection of inorganic bromine from tropospheric origin into the lower stratosphere (WMO report 2007, Chapter 2 on halogenated very short-lived substances VSLS). As a result, the ozone loss due to bromine might be underestimated in current models and in ozone trend simulations (e.g. Salawitch et al., 2005; Feng et al., 2007).

In the years 1980s, continued research led to the scientific consensus that man-made CFCs and halons posed a serious threat to the ozone layer. In response, the Montreal Protocol agreement was negotiated in 1987. This agreement regulated the production of chlorofluorocarbons and halons. Some revisions of this agreement have been made latter in the light of advances in scientific understanding. Due to its widespread adoption1 and adherence it has been hailed as an example of exceptional international cooperation and success. The main CFCs, CH3Br and halons were not produced anymore by any of the signatories after the end of 1995. The CFCs were replaced by partly halogenated substitutes (the so-called HCFCs), which are less stable and hence have a shorter lifetime.

The cumulative levels of chlorine are now decreasing in both the troposphere (Montzka et al., 1996) and the stratosphere (Anderson et al., 2000; Froidevaux et al., 2006). Long- term observations of CH3Br at the surface have shown that this substance, after peaking in 1998, has declined by 1.3 pptv by mid-2004 (Montzka et al., 2003; see also Figure 2.6 a). In case of the halons, they are still increasing, but at a slower rate. Decadal observations of stratospheric BrO from balloon (Dorf et al., 2006) and ground-based (Hendrick et al., 2008) instruments have confirmed that the decline observed in methyl bromide is now followed in a decline in stratospheric inorganic bromine. This is illustrated in Figure 2.6 where long-term observations of stratospheric BrO from ground- based instruments at two different sites, together with measurements of CH3Br and halons are shown. At both stations, a positive trend of BrO of about + 2.5% per year is found for the 1995-2001 period, while a negative trend of about -1% per year is obtained between 2001 and 2005. Note that there is a discrepancy between the moment the tropospheric organic bromine has peaked (~1998) and the moment the stratospheric bromine became maximum (~2001). This delay of approximately 3 to 4 years corresponds to the time for the organic bromine species emitted at the surface, to be transported into the stratosphere.

1 The Montreal Protocol and amendments have been signed by 193 countries.

Figure

Updating...

Références

  1. , “vapor
  2. , “sphere
Sujets connexes :